The problem with solid surface interactions with living biological tissue is a repeating subject within areas of medical technology, e.g. biomaterials, biosensors, and controlled drug delivery. Other application areas are food processing technology and biotechnical process chemistry, areas where wanted or unwanted interactions with biologically produced substances exist. Thus, there is a constant need for new materials with improved functions and characteristics, and there is an increasing need of experimental surface modifications suitable for the specific application areas.
The field of nanotechnology has made great progresses during the last decades, mainly due to the fact that nanostructured materials, with a structure size of 1-1000 nm, have very interesting properties with regards to optimized interaction with biological fluids and living tissue. Central to this patent application is a recently published paper describing a method of making nanostructured solid surfaces with nanostructures around 10 nm [1]. The method includes that flat gold surfaces are allowed to react with thiol terminated, lineary alkanes (dithiols), binding to the gold surface with one thiol group, whereas the other thiol end will constitute an overlayering carpet of pristine thiol groups.
A stable colloidal solution of negatively charged gold particles in the size range of 8-12 nm was brought into contact with the surface and the gold particles were adsorbed to the aforementioned pristine thiol groups. It was observed that the distance between the adsorbed particles could be controlled by varying the ionic strength of the citrate buffer that was used during the adsorption. The distance (center to center) could be varied between 10-100 nm when the molarity (concentration) of the buffer was varied between 10-0 nM as judged from visualization experiments with scanning electron microscopy (SEM). Similar results have earlier been demonstrated in particle adsorption experiments with electrostatically stabilized solutions of surface charged polymer particles to mineral surfaces such as glass, silicon dioxide or mica [2-5]. In addition, similar results have also been presented concerning adsorption of negatively charged gold surface nanoparticles from electrostatic stabilized solutions to glass surfaces or silicon dioxide surfaces that are positively charged due to chemical modifications [6].
The principle of electrostatically controlled particle adsorption is shown in FIG. 2. An electrostatically stabilized solution 202 containing surface charged nanoparticles 200 is applied to a beaker 201 (FIG. 2A). A surface preparation 203 is introduced in the container (FIG. 2B) allowing the particles 200 to bind to the surface 203 by means of electrostatic, semi-covalent, covalent or other types of bindings, leading to that a stable adsorption is obtained after a certain time (FIG. 2C). The particles have a certain distance, r, from each other.
This condition represents a terminal condition for the adsorption, and prolonged incubation time does not have any further impact on surface coverage of particles. When the surface is removed from the particle solution, the distance between two adjacent particles can be estimated from the interaction pair-potential according to the DLVO-theory [1,7], FIG. 3.
In short, the pair potential U(r), where r is the distance between two particles, may be calculated as the sum of an attractive potential Uattraction(r) emanating from the dispersive forces between the particles as well as a repulsive potential Urepulsion(r) emanating from the electrostatic repulsion between the particles.
The shape of the repulsive potential can be calculated in different ways, but will always vary with the so called Debye-distance which is an approximate measure of the declination of the potential outside the surface of the particle. A short Debye-distance means that the repulsive potential is rapidly declining outside the surface of the particle. The Debye distance is in turn dependent of the ionic strength in the particle solution 202, and can be expressed as:
      κ          -      1        =            [                                    ɛɛ            0                    ⁢          kT                          1000          ⁢                      e            2                    ⁢                      N            A                    ⁢          2          ⁢          I                    ]              1      /      2      
where ε is the relative permittivity, ε0 is the permittivity I in vacuum, k is the Boltzmann constant, T is the temperature, e is the elementary charge, NA is Avogadro's constant and the ionic strength is:
  I  =            1      2        ⁢                  ∑                  i          =          1                n            ⁢                        c          i                ⁢                  z          i          2                    
where ci and zi is the molar concentration and the charge of an i ion in the solution.
The Debye-distance, and the range of the repulsive potential, is therefore reduced with increasing ionic concentration in the colloidal solution 202. This means that each particle can bind closer to other particles on the surface when the condition
            U      ⁡              (        r        )              kT    =      1    λ  
where U(r) is the pair potential, kT is the thermal energy, and λ is a constant, is fulfilled for smaller r.
Surfaces prepared with dithiols and gold nanoparticles as described above and in [1, 8] have been used in biological experiments. In these experiments, the spaces between the particles were made protein repelling with a conjugated maleimide reagent which rapidly binds covalently to dithiol groups. The maleimides were conjugated with polyethylene glycol (PEG) which resulted in the spaces between the particles becoming repelling for proteins and cells. The surface on the absorbed gold particles could later be modified with thiol reagents, e.g. thiol with methyl groups, which gives the adsorbed gold particles hydrophobic characteristics. Surfaces with gold particles made in this way has a very highly controlled chemical structure and physical organization in the nanometer range which makes such surfaces well suited for adhesion studies of different kinds. The described method is very flexible for adhesion studies due to the relatively large number of commercial substances with malemide functions which can bind between the adsorbed particles, as well as thiol reagents which can bind to the adsorbed gold particles.
Similar experiments have been performed where gold nanoparticles, stabilized by polymers, have been applied to silica and glass surfaces by so called “dip-coat” technology [9]. Note that this method does not utilize electrostatic repulsion between particles to control their spread throughout the surface but instead the distance is defined by the polymer structures that surround the particles in solution. Interaction between these particles and the adsorbed surface is weak, why the particles after adsorption must be sintered in the substrate, a process in which also the surrounding polymers disappears from the surface. The surface around the gold particles then becomes the underlying silica substrate which can be modified with functional silanes, e.g. PEG-modified silanes, which makes this surface resistant to bioadhesion. The particles' surfaces can be modified with thiol reagents, e.g. thiol conjugated to so called RGD-peptides, an amino acid sequence which mediates cell interactions.
In this experiment it has also been described that polymer particles in an electrostatic stabilized solution is adsorbed to charged mineral surfaces and that the distance between the adsorbed particles has been controlled with electrostatic repulsion according to the above description [10]. The adsorbed surfaces either have a native net charge, or have been charged through chemical modification, e.g. with functional silanes. The bond between the surface and particles has primarily been of electrostatic nature. The surfaces with the adsorbed polymer particles have been used as a lithographic template with which the polymer particle covered parts of the surface have been transformed to isles of gold in the size range of 10-1000 nm surrounded by the substrate surface material. The surrounding substrate surface was then modified, e.g. with poly-L-lysine-PEG. Lysine is a positively charged polymer, which is adsorbed by negatively charged surfaces, and when conjugated with PEG, in certain cases make these surfaces resistant to bio adhesion. The gold surfaces can then be modified with thiol reagents, e.g. linear alkane thiols, which make the gold surfaces hydrophobic. To these hydrophobic surfaces proteins can be adsorbed, e.g. the protein laminin. Such surfaces have been used to study cell proliferation and surface interaction.
All of the above described technologies can be used to study the importance of a surface nanostructure for the adhesion process, and can be used as a platform for the design of materials with desired biological characteristics.
Most adhesion studies are performed on surfaces with a constant chemical setup. When studying the importance of one type of surface modification it is common practice to use several surface preparations in order to analyze adhesion phenomena independently. This procedure, however, is time and labor consuming since several surface preparations must be prepared for each series of experiments. In addition, the methodological errors of measurements can be relatively large which means that the interpretation of the intended study of adhesion phenomena can be either incorrect or overlooked.
A method to limit the methodological error and to reduce time spent to prepare surfaces is to create gradients in the chemical characteristics on a surface. One example of such a method is the so called “wettability gradient”, a surface which is hydrophobic in one end and hydrophilic in the other [11]. Between these endpoints the controlled and continuous gradient of chemical characteristics is found. This type of surface gradient significantly reduces time to prepare as well as the methodological error, and is often used in academic research [12-14].
Several methods to prepare continuous chemical gradients on a surface are known, one of them the well known diffusion method, FIG. 1. In this case the method of action is such that a reagent 001, e.g. methylchlorosilane is mixed in a solvent with high density 002, e.g. trichloroethyleneacetate (tri-) The mixture is then layered under a different solvent 003, e.g. xylene with low density. Between these layers there is a surface 004, e.g. glass on which a gradient will form. In time the solvents start to diffuse into each other where also the set of reagent 001 diffuses and binds to the surface 004. At a specific time of diffusion a bound gradient of hydrophobic methyl groups has occurred on the hydrophilic glass surface [11]. How much of the reagent that binds to the surface at a certain position, and therefore the hydrophobicity at this position is determined by the concentration of the set of reagents 001 on the surface at this position in combination with the time during which the surface has been exposed to the reagent solution. This means that the characteristics of the obtained gradient are determined through kinetic control.
To manufacture an even gradient in particle density with the above described method should be difficult, since the binding of nanoparticles from an electrostatic stabilized solution to a surface which binds these particles usually is a very fast process in relation to the particles rate of diffusion. This is low compared to the rate of diffusion for small molecules, such as methylchlorosilane. Attempts made to control the distance between nanoparticles on surfaces, where nanoparticles has been adsorbed to binding surfaces from electrostatic stabilized solutions, through varying the particle concentration and time of incubation, has shown the difficulties in controlling low density gradients of particles. Also, the particles do not show the same conformity of organization on the surface as they do after electrostatic controlled adsorption as described above [6, 15].
Recently, a gradient of gold particles on a silica substrate was disclosed, where the structuring of bound particles was good [16]. This gradient, described in [16] was manufactured according to a modified “dip-coat”-method, but without using electrostatic control or diffusion gradients. The obtained gradient had limited dynamic and the smallest particle distance was about 50 nm. The gradients were modified chemically with PEG between the particles and RGD peptides on top of the particles. The gradient surfaces were subsequently used in experiments to investigate cellular adhesion. In general, this publication discloses that the interest to make surface bound density gradients of gold particles is great, for reasons mentioned above. The technical solution to produce gradients according to [16] is however significantly more complex than the present invention.